Request for missing file Homo_sapiens.GRCh38.gene_symbol_to_embedding_ESM2.pt

[Zyx-WMU]

Hello @Raingate,

Thank you for sharing your work on MAP. I am trying to reproduce the knowledge encoder pre-training, but the script D_extract_se_emb.py (and later train_resume.py) fails because it cannot find the file Homo_sapiens.GRCh38.gene_symbol_to_embedding_ESM2.pt. This file is referenced in configs/se600m.yaml and model/model.py.

I have already downloaded the CSV data from Hugging Face, but this .pt file is not included. Could you please provide a download link for this file (or clarify how to generate it)? It would be very helpful for the community to fully replicate your results. Thank you!

[Raingate]

Hi! Thanks for reaching out. We are currently updating our repository to support better reproducibility. The Homo_sapiens.GRCh38.gene_symbol_to_embedding_ESM2.pt can be found in our newly updated Google Drive, which is https://drive.google.com/drive/folders/1cV0ZTk92PguKS2nyii6dLV0IfqoSDHsQ?usp=drive_link.